❌

Normal view

Team uses AlphaFold AI to redesign gene-editing proteins to make them safer

24 July 2026 at 18:31

A couple of decades after the discovery of systems that could selectively target DNA, we're starting to see the first therapies based on gene editing. One challenge these developments have faced is safety. While we can make them pretty specific to the gene we want edited, the human genome is very large, and even rare DNA sequences can appear a couple of times by chance.

As a result, all the original gene-editing systems had known rates of what are called off-target effects, in which they simply edit the wrong sequence. This may be a low-probability event, but edit enough cells—and therapies generally have to edit many—and errors become inevitable.

A lot of effort has gone into finding ways to minimize or eliminate off-target edits. In a recent issue of Nature, researchers described modifying the AI protein-folding software AlphaFold to help identify key areas of gene-editing proteins responsible for off-target effects. Those areas were then modified to reduce the problems.

Read full article

Comments

© Anton Vierietin

Team uses AlphaFold AI to redesign gene-editing proteins to make them safer

24 July 2026 at 18:31

A couple of decades after the discovery of systems that could selectively target DNA, we're starting to see the first therapies based on gene editing. One challenge these developments have faced is safety. While we can make them pretty specific to the gene we want edited, the human genome is very large, and even rare DNA sequences can appear a couple of times by chance.

As a result, all the original gene-editing systems had known rates of what are called off-target effects, in which they simply edit the wrong sequence. This may be a low-probability event, but edit enough cells—and therapies generally have to edit many—and errors become inevitable.

A lot of effort has gone into finding ways to minimize or eliminate off-target edits. In a recent issue of Nature, researchers described modifying the AI protein-folding software AlphaFold to help identify key areas of gene-editing proteins responsible for off-target effects. Those areas were then modified to reduce the problems.

Read full article

Comments

© Anton Vierietin

Adobe explora mais recursos de IA em seu app de fotografia Project Indigo

21 July 2026 at 14:39

A Adobe anunciou a adição de novos recursos temporários impulsionados por inteligência artificial (IA) ao Project Indigo, seu app de câmera exclusivo para iOS, com a ferramenta AI Playground.

Os quatro principais recursos de IA do Project Indigo incluem a manipulação de objetos (Object Editing, ou “edição de objeto”, em tradução direta), a estilização de uma foto em uma determinada técnica artística (Styles, ou “estilos”), o feedback por IA das suas fotos (Photo Guidance, ou “orientação fotográfica”) e a edição a partir de comandos de linguagem natural (Custom Edit, ou “edição customizável”).

Recurso Styles do Project Indigo da Adobe
Divulgação/Adobe

De forma opcional, a Adobe afirmou ainda que o acesso ao AI Playground estará restrito a uma parcela dos usuários do Project Indigo, sem menção aos critérios exatos de disponibilização. O teste ocorrerá durante algumas semanas e a ideia é compreender o uso de IA no app, embora ela reitere seus pilares de privacidade:

Quem aceitar participar precisará concordar em compartilhar apenas as interações com os botões do recurso. A Adobe afirma que não acessará os comandos digitados, nem as imagens, não fará upload delas para seus servidores e não as utilizará para treinar modelos de IA. Como não há login, todos os dados coletados são anônimos. Quem preferir não participar pode recusar e continuar usando o aplicativo normalmente.

Recursos como a estilização de fotos ou a remoção de objetos farão uso do modelo Nano Banana, do Google, ou dos próprios modelos Firefly da Adobe, mas ela reitera que o leque de modelos poderá se expandir no futuro.

Por fim, a Adobe explicou ainda que os modelos atuais de geração de imagens produzem apenas resultados com alcance dinâmico padrão (SDR 1Standard dynamic range, ou faixa dinâmica padrão.) e, para preservar a aparência das fotos capturadas com HDR 2High dynamic range, ou ampla faixa dinâmica. pelo Project Indigo, a empresa desenvolveu um modelo próprio de ML 3Machine learning, ou aprendizado de máquina. que converte automaticamente o resultado gerado pela IA para HDR, utilizando a foto original como referência, sem empregá-la para treinar modelos de IA.

via TechCrunch

Notas de rodapé

  • 1
    Standard dynamic range, ou faixa dinâmica padrão.
  • 2
    High dynamic range, ou ampla faixa dinâmica.
  • 3
    Machine learning, ou aprendizado de máquina.

Gene edit boosts rice safety while preserving high harvest yields

16 July 2026 at 13:15

Cadmium (Cd) contamination is a growing threat to food safety, and rice is one of the most vulnerable crops. Because rice plants can accumulate cadmium from contaminated soils more readily than many other staples, Cd can become a major dietary exposure route for large parts of the world’s population. Yet attempts to lower cadmium often create trade-offs—reducing uptake of essential nutrients or weakening plant performance.

A new study from Okayama University, Japan, and collaborating researchers in China reports a precision-editing strategy designed to avoid those trade-offs. The team focused on the rice metal transporter gene OsNramp5, which is known to move metals such as manganese and cadmium, and used base-editing to search for a beneficial point mutation within the gene rather than disabling it.

Using adenine and cytosine base editors, the researchers generated more than 1,600 genome-edited rice lines and screened them for reduced cadmium accumulation. Their saturation mutagenesis approach targeted OsNramp5 to identify variants that kept normal plant physiology while limiting Cd buildup in grain.

The key breakthrough was the I441T substitution: changing isoleucine to threonine at amino acid position 441. This single change reduced cadmium levels in both shoots and brown rice without altering gene expression, protein abundance, or the subcellular localization of the transporter.

Importantly, the mutation did not come at the cost of micronutrients. Field trials on cadmium-contaminated soil showed a 48% decrease in cadmium concentration in brown rice, from 0.14 mg/kg in wild type to 0.07 mg/kg in the edited plants, while iron, manganese, and zinc levels remained essentially unchanged.

The researchers then explored the mechanism behind the selective effect. Although OsNramp5 transports multiple metals, the I441T mutation shifted its selectivity—enhancing zinc transport preference. Increased zinc accumulation in root cells promoted competitive inhibition against cadmium during root-to-shoot movement.

Instead of completely blocking cadmium uptake, the edited transporter selectively limited cadmium translocation, reducing grain contamination while preserving essential mineral delivery. The team supported these conclusions with physiological assessments, gene and protein analyses, yeast transport assays, and agronomic evaluation.

Overall, the work demonstrates how precision genome editing can address a long-standing breeding challenge: lowering toxic metal accumulation without disrupting nutrient homeostasis. The researchers propose that the OsNramp5I441T allele could accelerate development of safer rice varieties for mildly contaminated regions.

Subject of Research: Not applicable
Article Title: Genome-edited rice variety with low-cadmium accumulation in the grain
News Publication Date: 18-Jun-2026
Web References: http://dx.doi.org/10.1073/pnas.2610609123
References: DOI: 10.1073/pnas.2610609123 (Proceedings of the National Academy of Sciences)
Image Credits: Credit: Professor Jian Feng Ma from Okayama University, Japan

Keywords: cadmium, rice, OsNramp5, base editing, genome editing, metal transporters, food safety, micronutrients, zinc competition, plant breeding

Programmable mRNA 3′UTR Restores MHC-I, Fights Prostate Cancer

16 June 2026 at 15:59

In a groundbreaking study poised to redefine cancer immunotherapy paradigms, researchers have unveiled a programmable RNA engineering platform that restores immune recognition in notoriously immune-resistant tumors such as prostate cancer. This innovative approach, centered on precise manipulation of mRNA alternative polyadenylation (APA), revives the expression of major histocompatibility complex class I (MHC-I) molecules — the pivotal sentinels of immune surveillance routinely downregulated in these tumors. The findings, detailed in Nature Biomedical Engineering, shed light on an elusive molecular mechanism of immune evasion and herald a new frontier in translational cancer immunology.

Prostate cancer and analogous immune-cold tumors have long bedeviled clinicians by resisting immune checkpoint inhibitor therapies, a frontline strategy in modern oncology. A primary culprit in this resistance is the diminished presentation of tumor-specific antigens on MHC-I molecules, molecules intrinsically responsible for flagging abnormal cells to cytotoxic CD8 T cells. Without adequate MHC-I display, these malignant cells evade immune detection, rendering checkpoint blockade largely ineffective. Despite the clinical significance, no therapies have yet emerged that selectively restore MHC-I expression, underscoring a critical unmet medical need.

The investigative team addressed this gap by engineering the 3′UTR CRISPR/dCas13 system (3′UTRCES), an advanced RNA-editing platform designed to modulate mRNA 3′ untranslated region (3′UTR) dynamics in vivo. Unlike conventional gene editing tools that alter DNA sequences, 3′UTRCES uniquely targets RNA transcripts’ polyadenylation sites to manipulate APA patterns — a post-transcriptional regulatory mechanism influencing mRNA stability, localization, and translation. Leveraging a catalytically inactive Cas13 (dCas13) fused to programmable effectors, this system reprograms 3′UTR architecture to reinstate normal gene expression profiles critical for immune function.

Central to their discovery is the identification of tumor-specific 3′UTR shortening in the E3 ubiquitin ligase adaptor SPSB1, a molecular player hitherto not linked to immune modulation. Through sophisticated analyses, the researchers demonstrated that aberrant SPSB1 3′UTR truncations promote enhanced ubiquitination and subsequent degradation of MHC-I molecules. This selective downregulation disrupts the immune system’s ability to recognize and eliminate malignant cells without affecting programmed death-ligand 1 (PD-L1), an immune checkpoint protein. These insights decode a novel mechanism by which tumors suppress antigen presentation, bypassing immune control.

Armed with this mechanistic understanding, the team deployed lipid nanoparticle (LNP) delivery of the 3′UTRCES constructs directly into syngeneic mouse models harboring prostate tumors. This therapeutic intervention reversed the aberrant SPSB1 3′UTR shortening, effectively restoring MHC-I expression on tumor cells. The restored antigen presentation reinvigorated the infiltration of CD8 cytotoxic T lymphocytes, amplifying antitumor immune responses and sensitizing tumors to subsequent immune checkpoint therapies. This combinatorial strategy outperformed existing modalities, offering a tangible path to overcome immune resistance.

The implications of this study extend beyond prostate cancer, representing a conceptual shift in targeting post-transcriptional gene regulation to modulate tumor immunity. By precisely tailoring RNA isoforms through 3′UTRCES, it becomes feasible to rectify pathological gene expression patterns without permanent genomic alterations, minimizing off-target effects and paving the way for tunable, reversible therapies. This technological stride potentially positions LNP-3′UTRCES as a versatile platform for broad-spectrum cancer immunotherapy.

Delving deeper into the molecular underpinnings, the research highlights how alternative polyadenylation — once considered a subtle nuance of gene regulation — plays an outsized role in immune evasion. APA events that truncate key regulatory 3′UTRs can rewire protein expression profiles, impacting ubiquitin-mediated degradation pathways. In SPSB1’s case, 3′UTR shortening shields MHC-I from stable expression, underscoring the delicate interplay between RNA processing and immune visibility. These revelations chart a new investigative avenue for uncovering APA-driven oncogenic mechanisms.

The study harnessed state-of-the-art transcriptomic and proteomic methodologies to validate the APA signatures and subsequent functional outcomes. Single-cell RNA sequencing and cross-linking immunoprecipitation unveiled the precise RNA-protein interactions modulated by 3′UTR alterations. Meanwhile, in vivo tumor growth assays meticulously quantified the therapeutic efficacy of 3′UTRCES-mediated intervention. The robust experimental design marries molecular biology with immuno-oncology, reinforcing the translational promise of RNA engineering.

Beyond prostate cancer, the researchers posit that similar APA-driven immune escape mechanisms may underpin resistance in other malignancies characterized as immune-cold or immunologically excluded. Tumor heterogeneity presents obstacles, yet the adaptability of the 3′UTRCES platform offers a pathway to customize RNA editing interventions tailored to tumor-specific APA landscapes. This precision medicine approach aligns with emerging oncology paradigms striving to unify genetic and epigenetic therapeutic targeting.

Crucially, this work confronts a longstanding conceptual bottleneck — the absence of clinically viable modalities to restore MHC-I expression selectively. Immune checkpoint inhibitors, though transformative, rely heavily on intact antigen presentation machinery to elicit durable responses. By restoring the tumor’s antigen-presenting capacity, 3′UTRCES synergizes with immune checkpoint blockade, reinvigorating exhausted cytotoxic T cells and promoting tumor clearance. This paradigm underscores the importance of targeting upstream immune suppression pathways in cancer.

The LNP delivery mechanism employed is equally noteworthy, showcasing advances in nanomedicine that enable precise nucleic acid transport to tumor sites. Lipid nanoparticles offer biocompatibility, customizable surface chemistry, and efficient endosomal escape — attributes critical for clinical translation. As the field gravitates toward RNA-based therapeutics post-COVID-19, leveraging these delivery systems in oncology sets a precedent for future innovations combining synthetic biology and immunotherapy.

Looking ahead, the scalability and safety of 3′UTRCES-mediated therapies warrant thorough exploration through preclinical and clinical studies. Immunogenicity of RNA-guided effectors, off-target RNA interactions, and durability of therapeutic responses remain important considerations. Nevertheless, the current findings provide a compelling proof of concept, invigorating translational research efforts to combat immune escape mechanisms effectively.

This pioneering research not only reframes our understanding of tumor immune evasion but also exemplifies how RNA technology can be harnessed to unlock therapeutic opportunities in recalcitrant cancers. By transforming APA from a molecular signature of immune suppression into a therapeutic target, the study opens vistas for next-generation cancer treatments that restore immune competence at the RNA level. This convergence of molecular biology, immunotherapy, and RNA engineering heralds a new dawn in oncology innovation.

In summary, the programmable editing of mRNA 3′UTRs via the 3′UTRCES platform revitalizes MHC-I expression in prostate tumors, counteracting immune suppression driven by SPSB1-mediated degradation pathways. This novel mechanistic insight, coupled with effective lipid nanoparticle delivery and demonstrated synergy with immune checkpoint inhibitors, marks a transformative advance in overcoming tumor immune resistance. The study exemplifies how precision RNA engineering can recalibrate tumor immunogenicity and redefine therapeutic potentials in cancer treatment.

As research continues to unravel the complexities of RNA-mediated regulation of immune visibility in cancer, the integration of programmable RNA tools like 3′UTRCES with established immunotherapies holds immense promise. These synergistic strategies may soon accelerate the development of personalized cancer therapies designed to circumvent immune escape and foster more durable clinical remissions. The future of oncology may indeed lie at the interface of RNA engineering and immunomodulation.

This landmark work not only elucidates a previously unrecognized molecular mechanism of immune evasion but also delivers a versatile, programmable therapeutic platform for precise post-transcriptional gene regulation in cancer. Its translational implications inspire optimism for expanding the repertoire of RNA-based interventions that can be tailored to diverse tumor contexts, ultimately enhancing patient outcomes and transforming cancer immunotherapy at its molecular roots.


Subject of Research: Programmable RNA engineering to restore MHC-I expression and overcome immune evasion in prostate cancer.

Article Title: Programmable mRNA 3′UTR engineering restores MHC-I and overcomes immune evasion in prostate cancer.

Article References:
Huang, F., Yuan, F., Li, K. et al. Programmable mRNA 3′UTR engineering restores MHC-I and overcomes immune evasion in prostate cancer. Nat. Biomed. Eng (2026). https://doi.org/10.1038/s41551-026-01720-9

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41551-026-01720-9

❌